Bridging the gap between complex payment orchestration, D2C eCommerce, and Production AI Systems.

Currently driving technical solutions at Juspay (Aug 2024 - Present).

Languages & Core

AI/ML & Data Engineering

Cloud & DevOps Infrastructure

Production-grade evaluation framework for "Build vs Buy" AI decisions.

• Architecture: Automated suite benchmarking GLM-4, Mistral, and QwQ models using SSH tunneling.
• Key Achievement: Developed a pipeline to measure Token-Per-Second (TPS) latency, memory footprint, and response quality to optimize inference costs.

High-performance payment orchestration for D2C scale.

• Role: Technical implementation for 125+ brands, handling dynamic payment routing and A/B testing on checkout UI.
• Key Achievement: Automated 30% of engineering dependency via custom playbooks and reduced incident resolution time by 35%.

Accessible communication tool for the hearing impaired.

• Architecture: Two-stage pipeline using CNN (palm detection) + RNN (temporal gesture recognition).
• Optimization: Hybrid model quantization for low-latency inference on mid-range edge devices.

Offline-first mental wellness application.

• Architecture: Hybrid data sync architecture with custom conflict resolution logic for intermittent connectivity.
• Stats: 1K+ Downloads on Play Store.

I believe in mastering the tools to maximize efficiency.

• 💻 Terminal: tmux for multi-session workflows + bash automation.
• ☁️ Cloud: gcloud compute ssh with IAP tunneling for secure VM management.
• 📝 Editor: VS Code + Cline for AI-assisted coding.